# Black Forest Labs Launches FLUX 3 Capable of Generating Images and 20-Second Videos The new multimodal model FLUX 3 is now available in limited release. - **Source:** [VentureBeat (feeds.feedburner.com)](https://brifing.si/source/c51474dc-6b09-49dd-aaca-085f4ed00c84) - **Published:** 2026-07-23 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/241c9d8e-5ef3-44b8-b544-e452c9b686b5 ## Launch of FLUX 3 Black Forest Labs has expanded its FLUX family with the announcement of FLUX 3, a new model capable of generating images, 20-second videos, and audio under a single architecture. The company refers to these capabilities as 'visual intelligence,' emphasizing perception, prediction, and action across physical and digital environments. ## Multimodal Capabilities FLUX 3 is trained to generate images, video, and audio within a unified architecture, rather than combining separate models. This approach ensures more consistent and integrated outputs. The company highlights potential applications in creative generation, simulation, and robotics. ## Product Lines FLUX 3 will be offered through four product lines: FLUX 3 Video, FLUX 3 Image, FLUX 3 Action, and the upcoming open-source FLUX 3 Dev. FLUX 3 Video and FLUX 3 Action are currently in an early access program, requiring approval for use. FLUX 3 Image is set to roll out in the coming weeks. ## Limited Availability FLUX 3 is initially available in limited release, with no public API access yet. This strategy mirrors other leading AI labs. The company has promised faster and open-weight versions later this year. ## Pricing Uncertainty BFL has not disclosed pricing or service-level commitments for FLUX 3, making it difficult for enterprise buyers to calculate total cost of ownership. Independent benchmarking data is also lacking. ## Lack of Open Source FLUX 3 is launching without downloadable weights or an open-source license, disappointing developers accustomed to locally deployable FLUX variants. The company assures open-weight versions are coming. ## Comparative Testing BFL claims FLUX 3 outperforms competitors in early tests, with 93% preference over Luma Ray 3.2 and 77% over Runway Gen-4.5. However, these results are preliminary and may not reflect the final model's performance. ## Competition with Google Gemini Omni FLUX 3 ties with Google's Gemini Omni Flash at 52% in quality comparisons. Omni is generally available at $0.10 per second for 720p video, but has editing restrictions in Europe, potentially giving FLUX 3 an edge. ## Opportunity in the European Market As a Germany-based company, BFL may have an advantage in Europe, where Google Omni faces restrictions on editing uploaded videos. This could make FLUX 3 more appealing to European enterprises. ## Enterprise Guide FLUX 3 Video supports single generations up to 20 seconds, but resolution and pricing details remain undisclosed. Competitors like HappyHorse 1.1 offer 15-second 1080p clips, leaving questions about FLUX 3's competitive positioning. ## Future Commitments BFL promises FLUX 3 Dev will provide open-weight access to a multimodal backbone for content creation and action prediction, a significant expansion over previous FLUX releases. However, developers will need to wait for this version. ## Conclusion and Significance FLUX 3 represents a potential leap in multimodal AI, but limited availability and missing details may slow enterprise adoption. Open-source availability and pricing transparency will be critical to its success. ## Sources covering this story - https://venturebeat.com/technology/black-forest-labs-launches-flux-3-capable-of-generating-images-and-20-second-video-with-audio-but-in-limited-release-to-start — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._